Mercurial > hg > machine-learning-hw1
changeset 13:f07e20181cf0 default tip
Change step size in ex1_multi.m and renormalise new inputs for gradientDescentx
| author | Jordi Gutiérrez Hermoso <jordigh@octave.org> |
|---|---|
| date | Wed, 26 Oct 2011 07:57:12 -0700 |
| parents | 2af1401f7f7d |
| children | |
| files | ex1_multi.m |
| diffstat | 1 files changed, 5 insertions(+), 4 deletions(-) [+] |
line wrap: on
line diff
--- a/ex1_multi.m +++ b/ex1_multi.m @@ -82,8 +82,8 @@ fprintf('Running gradient descent ...\n'); % Choose some alpha value -alpha = 0.01; -num_iters = 100; +alpha = 0.03; +num_iters = 1000; % Init Theta and Run Gradient Descent theta = zeros(3, 1); @@ -104,7 +104,8 @@ % ====================== YOUR CODE HERE ====================== % Recall that the first column of X is all-ones. Thus, it does % not need to be normalized. -price = 0; % You should change this +Price_norm = bsxfun (@rdivide, bsxfun(@minus, [1650 3], mu), sigma); +price = [1 Price_norm]*theta; % ============================================================ @@ -149,7 +150,7 @@ % Estimate the price of a 1650 sq-ft, 3 br house % ====================== YOUR CODE HERE ====================== -price = 0; % You should change this +price = [1 1650 3]*theta; % You should change this % ============================================================